Synthesis, Characterization, and Catalytic Properties of Iridium Pincer Complexes Containing NH Linkers

Article abstract of DOI:10.1021/acs.organomet.7b00713

A series of tert-butyl-substituted pincer ligands based on 1,3-diaminobenzene and 3-aminophenol scaffolds, ^tBu4PXCYP (1e, X = Y = NH; 1f, X = NH; Y = O) and the corresponding iridium hydridochloro complexes (^tBu4PXCYP)IrHCl (2e, X = Y = NH; 2f, X = NH; Y = O) were prepared with moderate yields and high purity and were fully characterized by ¹H and ³¹P NMR spectroscopy. Unsymmetrical hybrid pincer ligands ^R2PNCOP^tBu2 (1g, R = isopropyl; 1h, R = cyclohexyl) were prepared conveniently in high yield via a one-pot procedure by judiciously choosing reaction conditions and base; the corresponding iridium hydrido chloro complexes ^iPr2PNCOP^tBu2IrHCl (2g) and ^Cy2PNCOP^tBu2IrHCl (2h) were synthesized by the reaction of [IrCl(COE)₂]₂ with ligands. X-ray crystallography reveals that these iridium pincer complexes adopt similar square-pyramidal geometries and exhibit strong intermolecular hydrogen bonding between the NH linker and chloride ions of the adjacent iridium complex in the solid state. ¹H NMR chemical shifts of tert-butyl based pincer ligated iridium hydrides move downfield when the electronegativity of the linker between the benzene backbone and phosphine moiety increases for 2a, 2e, 2f, and 2b. Accordingly the corresponding iridium pincer carbonyl complexes (^tBu4PXCYP)Ir(CO), 3a, 3e, 3f, and 3b show a blue shift in the CO stretching frequency. The activities of iridium complexes containing NH linkers were briefly examined for transfer dehydrogenation from cyclooctane to tert-butylethylene; (^iPr2PNCOP^tBu2)IrHCl (2g) exhibits the highest activity among all tested iridium pincer complexes, including the most studied (^tBu4PCP)IrHCl (2a) and (^tBu4POCOP)IrHCl (2b). The enhanced catalytic activity could be related to combined electronic and steric effects of the NH/O hybrid linker and different alkyl groups at phosphorus. This new class of iridium pincer complexes could have great implications in catalytic transformation of polar compounds due to the strong hydrogen-bond-donating ability of the NH linker.

Full text of DOI:10.1021/acs.organomet.7b00713

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Synthesis, Characterization, and Catalytic Properties of Iridium
Pincer Complexes Containing NH Linkers
Reuben B. Leveson-Gower,† Paul B. Webb,† David B. Cordes,‡ Alexandra M. Z. Slawin,‡
David M. Smith,† Robert P. Tooze,† and Jianke Liu*,†
†Sasol Technology (U.K.) Ltd., Purdie Building, North Haugh, St. Andrews KY16 9ST, U.K.
‡School of Chemistry, University of St. Andrews, Purdie Building, North Haugh, St. Andrews KY16 9ST, U.K.
S
* Supporting Information
ABSTRACT: A series of tert-butyl-substituted pincer ligands based on 1,3-
diaminobenzene and 3-aminophenol scaffolds, tBu4PXCYP (1e, X = Y = NH;
1f, X = NH; Y = O) and the corresponding iridium hydridochloro complexes
(
tBu4PXCYP)IrHCl (2e, X = Y = NH; 2f, X = NH; Y = O) were prepared with
1
moderate yields and high purity and were fully characterized by H and 31P
NMR spectroscopy. Unsymmetrical hybrid pincer ligands R2PNCOPtBu2 (1g,
R = isopropyl; 1h, R = cyclohexyl) were prepared conveniently in high yield
via a one-pot procedure by judiciously choosing reaction conditions and base;
the corresponding iridium hydrido chloro complexes iPr2PNCOPtBu2IrHCl
(2g) and Cy2PNCOPtBu2IrHCl (2h) were synthesized by the reaction of [IrCl(COE)2]2 with ligands. X-ray crystallography
reveals that these iridium pincer complexes adopt similar square-pyramidal geometries and exhibit strong intermolecular
hydrogen bonding between the NH linker and chloride ions of the adjacent iridium complex in the solid state. 1H NMR chemical
shifts of tert-butyl based pincer ligated iridium hydrides move downfield when the electronegativity of the linker between the
benzene backbone and phosphine moiety increases for 2a, 2e, 2f, and 2b. Accordingly the corresponding iridium pincer carbonyl
complexes (tBu4PXCYP)Ir(CO), 3a, 3e, 3f, and 3b show a blue shift in the CO stretching frequency. The activities of iridium
complexes containing NH linkers were briefly examined for transfer dehydrogenation from cyclooctane to tert-butylethylene;
(
(
iPr2PNCOPtBu2)IrHCl (2g) exhibits the highest activity among all tested iridium pincer complexes, including the most studied
tBu4PCP)IrHCl (2a) and (tBu4POCOP)IrHCl (2b). The enhanced catalytic activity could be related to combined electronic and
steric effects of the NH/O hybrid linker and different alkyl groups at phosphorus. This new class of iridium pincer complexes
could have great implications in catalytic transformation of polar compounds due to the strong hydrogen-bond-donating ability
of the NH linker.
Chart 1. Iridium Pincer Complexes for Alkane
Dehydrogenation
INTRODUCTION
■
Catalytic alkane dehydrogenation has been extensively studied
for a few decades due to its potential importance in the
production of linear olefins as versatile chemical intermediates
from abundant alkane feedstocks. Pt-based heterogeneous
catalysts are widely used in commercial processes to
dehydrogenate alkanes to the corresponding alkenes under
harsh reaction conditions. Low conversion per pass is essential
to maintain high olefin selectivity; thus, alkane dehydrogenation
is only operated as an integrated part of the linear alkyl benzene
(LAB) production unit to circumvent technical difficulties in
product separation. Seminal work by Jensen et al.1 on
homogeneous iridium pincer complex (2a, Chart 1) catalyzed
alkane dehydrogenation opened up a new avenue for selective
functionalization of alkanes. Although other transition metals
(Rh, Ru, Os)2 and phosphine-free pincer ligands such as
Phebox3 have also been explored for alkane dehydrogenation,
iridium pincer complexes containing phosphines are the most
investigated class due to the great steric and electronic
tunability of the pincer ligand framework4−6 and high
thermodynamic driving force for C−H activation by iridium.7
A significant amount of work has then been dedicated to tuning
the steric properties of pincer ligands by altering alkyl groups at
the phosphorus atom of Jensen’s PCP system.8 For example,
Received: September 19, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Goldman et al.9 systematically varied the steric hindrance of
R4PCPIr complexes by sequential replacement of tert-butyl
groups with methyl groups, and (tBu3MePCP)IrH4 was found
more active than other variations. Substituting adamantyl for
tert-butyl significantly improved the thermal stability of iridium
pincer complexes,10 while the activity for alkane dehydrogen-
ation is comparable for (Ad4PCP)IrH2 (Ad = adamantyl) and
tBu4PCPIrH2. Brookhart et al. took a different approach and
investigated the effect of the linker between benzene backbone
and phosphorus atom of ligands on catalytic alkane
dehydrogenation activity.11 Substitution of CH2 linkers with
oxygen afforded the iridium bis(phosphinite) pincer complexes
(RPOCOP)Ir, which proved to be highly effective catalysts for
transfer dehydrogenation of alkanes. The POCOP-ligated
iridium complexes are air stable both in the solid state and in
solution and are less inhibited by olefin hydrogen acceptor or
product; these characteristics are of paramount importance for
bulk chemical production. More interestingly, catalytic activity
was improved significantly when only one CH2 linker was
replaced with oxygen; the resulting “hybrid” phosphine−
phosphinite pincer iridium complex (tBu4PCOP)Ir (2c) is 4
times more active than its bis(phosphine) analogue (tBu4PCP)Ir
(2a) and 8 times more active than its bis(phosphinite)
analogue (tBu4POCOP)Ir (2b) in the metathesis of n-hexane.12
Among various permutations of linkers and alkyl substitutents,
Chart 2. Pincer Ligands Containing NH Linker(s)
Synthesized in This Work
to assess experimentally the effect of O, CH2, S, and NH linkers
on the steric and electronic properties of a complete series of
tert-butyl-substituted iridium pincer complexes. In comparison
to benzene-based pincer ligands with CH2, O, and S linkers,
pincer ligands containing NH linkers could exhibit a very rich
chemistry, since NH in close proximity to a metal center will be
acidic and can function as a hydrogen bond donor to facilitate
polar bond activation. A promising catalytic performance of
iridium pincer complexes bearing NH linkers in dehydrogen-
ation of alcohols and hydrogenation of carbonyl compounds
can be anticipated. From the perspective of ligand design, the
NH moiety also introduces an additional anchor point for
further ligand modification to enable precise tuning of the steric
crowding about the metal center. Hence, the NH linker will
open new avenues for the rational design of new pincer ligands
for a wide range of applications.
(
tBu2PCOPiPr2)Ir was found to be the most active. More
recently, Huang et al.13 synthesized a number of iridium
complexes supported by hybrid phosphinothio/phosphinite
pincer ligands, among which (iPr4POCSP)Ir was found to be a
highly active catalyst for transfer dehydrogenation of n-alkanes
with high regioselectivity for α-olefin formation while the tert-
butyl substituted analogue (tBu4POCSP)Ir (2d) is much less
active most likely due to its steric bulkiness.14 Iridium pincer
complexes containing various linkers reported in the literature
are summarized in Chart 1.
In this study, we aimed to develop a simple modular
procedure to synthesize a series of 3-aminophenol-based
t
unsymmetrical pincer ligands tBu2POCNPR2 (R = Bu, 1f; R =
iPr, 1g; R = cyclohexyl (Cy), 1h) by independently varying the
linker, the alkyl substituents at phosphorus, or both. These
novel iridium complexes with symmetric and nonsymmetric
pincer ligands, 2e−h, were then prepared and fully charac-
terized by 1H, 31P{1H}, 13C{1H} NMR spectroscopy and X-ray
crystallography. The influence of the NH linker on the
electronic properties of iridium pincer complexes was probed
via the CO stretching frequency of the corresponding iridium
pincer carbonyl. Initial catalytic tests were also carried out to
compare newly prepared iridium pincer complexes with the
widely studied (tBu4PCP)Ir (2a) and (tBu4POCOP)Ir (2b) for
benchmark transfer dehydrogenation of cyclooctane (COA)
with tert-butylethylene (TBE).
Significant advances in synthesis and extensive applications of
phosphine-based iridium pincer complexes in catalytic alkane
dehydrogenation reactions have been reviewed periodically by
Goldman,15,16 Brookhart,17 and Huang.18 Encouraged by
promising catalytic activities of iridium complexes achieved by
systematically tuning both the linker between the benzene
backbone and the phosphine moieties and the alkyl substituents
at phosphorus, we turned our attention to investigate the
potential influence of NH linkers on the catalytic performance
of iridium complexes, which has been much less explored for
iridium pincers in the literature. In this work we prepared a
series of pincer ligands containing NH linkers (Chart 2) and
investigated the effect of the NH moiety on the electronic,
steric, and catalytic properties of the corresponding iridium
complexes.
RESULTS AND DISCUSSION
■
Synthesis of Pincer Ligands. The pincer ligands 1e,f
containing NH linkers are conveniently prepared by
diphosphorylation of 1,3-diaminobenzene or 3-aminophenol,
respectively, with di-tert-butylchlorophosphine (Scheme 1).
The strength of the base was found to be a crucial factor for
successful ligand synthesis. Sodium hydride was used as a base
for the synthesis of tBu4POCOP (1b) in the literature; a
sufficiently stronger base such as n-butyllithium is required in
this study to deprotonate amino functional groups in 1,3-
diaminobenzene and 3-aminophenol because amino groups
(pKa(anline) = 30.7)22 are much less acidic than the hydroxyl
groups of resorcinol (pKa = 9.15).23 It is worth highlighting that
triethylamine and n-butyllithium were added sequentially for
the preparation of 1e in the literature,19 while n-butyllithium
Although the synthesis of the 1,3-diaminobenzene-based
pincer ligand 1e and the corresponding square-planar
complexes of palladium, nickel, platinum,19 and cobalt20 have
been reported previously by Kirchner et al., the iridium
complex of 1e has not been reported so far to the best of our
knowledge. However, the effect of the linker on the electronic
properties of a series of (tBu4PXCXP)Ir (X = O, NH, CH2, S,
etc.) complexes has been investigated computationally.21 The
calculated reaction and activation energies for methane C−H
addition are very similar for (tBu4PXCXP)Ir (X = NH, 2e) and
(
tBu4POCOP)Ir (2b), implying the potential of an iridium
pincer complex containing an NH linker in alkane activation.
The synthesis and structural characterization of 2e will allow us
B
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and could be applicable to the synthesis of other pincer ligands
with similar structures. The two steps are believed to proceed
via different reaction mechanisms due to the different steric
bulk of chlorodialkylphosphines and the strength of the base
used, as proposed in Scheme 3. Direct deprotonation of the
Scheme 1. Synthesis of tert-Butylphosphorus-Substituted
Pincer Ligands 1e,f
Scheme 3. Proposed Reaction Mechanism for One-Pot
Synthesis of Unsymmetrical Hybrid Ligands
alone is effective in our synthesis. The pincer ligand 1e was
obtained in high yield; its 31P{1H} NMR spectrum displays a
characteristic singlet at 57.8 ppm. The tert-butyl-substituted
aminophosphine-phosphinite hybrid pincer ligand 1f was
obtained in lower yield and exhibits two singlets at 58.7 and
152.4 ppm, corresponding to the aminophosphine and
phosphinite moieties of the NH and O arms, respectively.
The isopropyl-substituted analogue of 1f was prepared
previously by Ozerov et al.24 using a similar procedure and
displays singlets at 48.9 and 145.5 ppm in the 31P{1H} NMR
spectrum.
hydroxyl group of 3-aminophenol with a strong base such as
sodium hydride in the first step forms the corresponding
sodium phenoxide as a strong nucleophile. This subsequently
attacks di-tert-butylchlorophosphine to afford a monophos-
phorylated intermediate, while the amino group remains intact
due to its low acidity and low nucleophilicity in comparison to
the phenoxide anion. In the second step, substitution of
diisopropylphosphine at the amino nitrogen atom might
proceed via an ammonium salt intermediate, as shown in
Scheme 3. It is proposed that the lower steric bulk of the
isopropyl groups (in comparison to tert-butyl groups) allows
for attack by amine as a relatively weaker nucleophile.
Conversely, anionic arylamide or phenoxide nucleophiles are
required for introduction of di-tert-butylphosphine at amino
nitrogen and alcohol oxygen atoms, respectively. The acidic
ammonium salt intermediate is then deprotonated by a
relatively weak base such as triethylamine to afford 1g as the
final product. Such a reaction pathway appears to be limited to
less sterically bulky dialkylphosphines, and a slight excess of di-
tert-butylchlorophosphine introduced in the first step does not
affect the second step and can be removed under reduced
pressure at the end of the synthesis.
We also extended our synthesis to prepare unsymmetrical
PNCOP hybrid pincer ligands 1g,h through the modular
approach shown in Scheme 2.
Scheme 2. One-Pot, Two-Step Synthesis of Unsymmetrical
Hybrid Pincer Ligands 1g,h
By judiciously choosing reaction conditions and base, 1g,h
were conveniently obtained in high yield and high purity from a
one-pot, two-step synthesis. In the first step, the hydroxyl group
of 3-aminophenol was selectively deprotonated by 1 equiv of
sodium hydride; subsequent addition of 1 equiv of di-tert-
butylchlorophosphine afforded a monophosphorylated inter-
mediate. In the second step, the intermediate thus obtained was
phosphorylated (without isolation and purification) using
diisopropylchlorophosphine with triethylamine as a base.
Ligand 1g was obtained in high yield. When dicyclohexyl-
chlorophosphine was used in the second step, the correspond-
ing pincer ligand 1h was obtained in high yield.
Huang et al. previously synthesized a series of PSCOP pincer
ligands with varying alkyl substituents at the S and O arms.14
However, the monophosphorylated intermediate was separated
and purified prior to the introduction of the second
dialkylchlorophosphine. The one-pot, two-step synthesis
reported here presents obvious advantages due to its simplicity
Synthesis of Iridium Complexes. The crude pincer
ligands 1e−h prepared above were used without any further
purification, and their reaction with [IrCl(COE)2]2 in refluxing
toluene yields the corresponding iridium pincer hydridochloro
complexes 2e,f with high purity in moderate yields (Scheme 4).
The most widely studied iridium pincer complexes, (tBuPCP)-
IrHCl (2a) and (tBuPOCOP)IrHCl (2b), were also prepared as
reference materials, according to the literature procedure.11
The 1H NMR spectra of (tBu4PNCNPIr)HCl (2e) and
tBu4PNCOPIrHCl (2f) show overlapping doublets or triplets
from the tert-butyl groups in the narrow range of 1.15−1.3 ppm
and triplets from the hydridic IrH moieties at −42.87 (2e) and
2
−42.13 ppm (2f) with JPH = 13.0 Hz. The unsymmetrical
hybrid pincer iridium complexes (iPr2PNCOPtBu2)IrHCl (2g)
and Cy2PNCOPtBu2IrHCl (2h) also display characteristic IrH
triplets at −40.85 and −39.89 ppm, respectively. 31P{1H} NMR
C
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
induced heavy atom effects on the hydride shifts, as revealed by
DFT calculations.25 The SOC effect is particularly significant
for iridium hydride complexes and is thought to be
predominantly responsible for the unusually high field 1H
chemical shifts observed. In contrast, the electronic/steric
influence of linkers and alkyl substituents at the phosphorus
atom, which are removed from the hydride ligand, become
minuscule. Other factors such as π conjugation between the p
orbital of linker atoms with lone pairs and the molecular
orbitals of the benzene ring also cannot be neglected. The
electronegative linker atoms could have two opposite effects:
on one hand, they withdraw electron density from phosphorus
and reduce the electron-donating abilities of phosphine ligands;
on the other hand, they could direct electron flow through π
conjugation with the benzene ring and ultimately donate at
least part of the electron density originating from the
phosphine back to the metal center. The net effect of the
linker groups on the electronic properties of the metal center is
thus very complex and cannot be solely explained by a single
spectral descriptor such as the chemical shifts of iridium
hydrides. Nevertheless, iridium pincer hydrido chloro com-
plexes generally serve as catalyst precursors for alkane
dehydrogenation, and so it is more preferable to investigate
the electronic properties of 14-electron, 3-coordinate iridium
pincer intermediates believed to be more relevant for alkane
dehydrogenation under the reaction conditions.17
CO Stretching Vibration as a Probe of the Structure of
Iridium Pincer Complexes. The CO stretching frequency is
widely used to determine the small changes in electron density
at the metal center of transition-metal carbonyl complexes.26
Brookhart et al. demonstrated that the shifts of CO stretching
frequency (νCO) of iridium pincer carbonyl complexes are well
correlated with the electronic properties of para substituents at
the benzene backbone27 and Lewis acid−base interactions with
the γ-alumina support for immobilized iridium catalysts.28 To
probe the electronic properties of catalytically relevant iridium
centers, (tBu4PNCNP)Ir(CO) (3e) and (tBu4PNCOP)Ir(CO)
(3f) were prepared in almost quantitative yield using modified
literature procedures,27 as shown in Scheme 5.
Scheme 4. Syntheses of Pincer-Ligated Iridium Hydrido
Chloride Complexes
spectra of the hybrid PNCOP iridium pincer complexes 2f−h
2
present characteristic doublets with JPP = 362 Hz from the
coupling between the two magnetically inequivalent phospho-
rus atoms. The iridium pincer hydridochlorides with NH
linker(s) synthesized here, 2e−h, have spectroscopic features
similar to those reported previously for 2a−d containing O and
S linkers.12,14 The H NMR chemical shifts of these iridium
1
hydrides are compared in Table 1.
The hydride NMR chemical shifts give some indication of
the electronic properties of the metal center and are influenced
by the other ligands. For the iridium complexes with
symmetrical linkers between the benzene ring and phosphorus
atoms, the chemical shift of the iridium hydride resonance
moves downfield from −43.48 ppm (2a) to −42.87 ppm (2e)
and −41.4 ppm (2b). Ostensibly, this trend indicates
decreasing electron density at the iridium metal centers of 2a,
2e, and 2b, which is in good agreement with increasing
electronegativity of the linker atom: C (2.55) < N (3.04) < O
(3.44). However, the iridium pincer hydrides with mixed linker
atoms between the benzene ring and phosphorus donor atoms
Scheme 5. Syntheses of Pincer-Ligated Iridium
Monocarbonyl Complexes 3e,f
(
(
tBu4PCOP)IrHCl (2c), (tBu4PSCOP)IrHCl (2d), and
tBu4PNCOP)IrHCl (2f) exhibit hydride resonances in the
narrow range of −41.38 to −42.13 ppm, showing no correlation
with the electronegativity of different combinations of CH2, S,
1
NH, and O linkers. Furthermore, H NMR chemical shifts of
the complexes are apparently affected by the bulkiness of alkyl
substituents on the phosphorus atoms. For the series of
unsymmetrical PNCOP iridium complexes, a hydride reso-
nance is present at −42.13 ppm for (tBu4PNCOP)IrHCl (2f),
while it shifts downfield to −40.85 and −39.89 ppm for
(
iPr2PNCOPtBu2)IrHCl (2g) and (Cy2PNCOPtBu2)IrHCl (2h),
respectively. Apparently, the hydride chemical shift is not a
reliable measure for the estimation of the electronic properties
of the metal center in these iridium pincer complexes. One
possible explanation could involve spin−orbit coupling (SOC)
1
Table 1. H NMR Chemical Shifts of Iridium Hydrides Supported by Pincer Ligands with Varying Linkers
2a
2e
NH
2b
2c
2d
2f
2g
2h
linker (−X−)
δH(IrH)/ppm
2JPH/Hz
CH2
O
CH2/O
−41.38
13.3, 12.3
S/O
−41.5
14
NH/O
−42.13
13.1
NH/O
−40.85
13.2
NH/O
−39.89
13.1
−43.48
12.6
−42.87
13.0
−41.40
13.1
D
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
The νCO stretching frequencies of the resulting carbonyl
complexes 3e,f are compared in Table 2 with those of tBu4PCP
(3a) and tBu4POCOP (3b) iridium carbonyls reported
previously.
Table 2. CO Stretching Frequencies of Iridium Pincer
Carbonyl Complexes
complex
νCO (cm−1
)
conditions
ref
tBu4PCPIr(CO) (3a)
1913
DCM soln
C6H6 soln
pentane soln
pentane soln
DCM soln
DCM soln
DCM soln
29
9
1917
1927.7
1949
27
27
30
tBu4POCOPIr(CO) (3b)
1937
tBu4PNCNPIr(CO) (3e)
tBu4PNCOPIr(CO) (3f)
1919.5
1928.2
this work
this work
As shown in Table 2, νCO stretching frequencies of the
iridium pincer carbonyl complexes vary depending on the
solvent in which the infrared spectra were recorded. The IR
spectra of 3e,f were recorded in dichloromethane (DCM) due
to their high solubilities in this solvent. IR spectra of 3a, 3e, and
3b in DCM show a blue shift of the νCO stretching frequencies
with increasing electronegativity of the linkers: from 1913 cm−1
for 3a (X = CH2) to 1919.5 cm−1 for 3e (X = NH) and 1937
cm−1 for 3b (X = O). The blue shift of νCO is consistent with
the decreasing electron-donating properties of iridium pincer
fragments in the order 3a > 3e > 3b, as would be expected. The
iridium pincer carbonyl with an NH/O hybrid linker (3f)
displays a CO absorption band at 1928.2 cm−1, which lies
between those of 3e and 3b, indicating that the effect of the
linker on the electronic properties of the metal center might be
additive. It is interesting to note that varying the linker between
the benzene backbone and phosphorus donor atom is more
effective in tuning the electronic properties of metal center than
altering alkyl substituents at phosphorus atoms. In this work,
νCO stretching frequencies of (tBu4PXCXP)Ir(CO) can be tuned
over a much wider range from 1913 to 1937 cm−1, when the
linker X is changed from CH2 to O and NH. In contrast, νCO
stretching frequencies of (R4PCP)Ir(CO) in hexane solutions
vary within a narrow range when the alkyl substituent R is
Figure 1. Thermal ellipsoid plots of compounds 2e−h with thermal
ellipsoids at the 50% probability level. Solvent molecules, as well as
hydrogen atoms bound to carbon, are omitted for clarity, as is the
second independent molecule in 2g.
As can be seen from Figure 1, all four new iridium pincer
complexes synthesized adopt a square-pyramidal geometry with
the hydride at the apical site with a vacant position trans to the
hydride. Selected bond lengths and bond angles of 2e−h are
compared with those of previously reported iridium pincer
complexes 2a−d in Table 3.
t
i
changed from Bu (1914 cm−1) to Pr (1918 cm−1) and to
adamantyl (1916 cm−1).10 The CO stretching frequency is
slightly more sensitive for (R4POCOP)Ir(CO) complexes; a red
shift of 7 cm−1 from (iPr4POCOP)Ir(CO) (1944 cm−1) to
The Ir−C bond lengths are almost identical for all iridium
pincer complexes and fall in the range of 2.011−2.019 Å. A
closer inspection reveals the structural difference between the
newly synthesized 2e,f and the well-studied 2a,b. The structures
of 2a,b are almost perfectly square pyramidal; the C1−Ir−Cl1
angles of 2a (179.7°) and 2b (179.11°) are very close to
linearity.32 New iridium pincer complexes prepared in this work
exhibit large distortion angles C1−Ir−Cl1 of 177.22, 178.8,
173.91. and 170.88° for 2e−h, respectively. Such large
distortion from a perfect square-pyramidal geometry could be
attributed to electronic effects of the NH and/or O linkers
(2e,f), steric effects of alkyl substituents at the NH linker (2f−
h), or both. However, extensive hydrogen-bonding interactions
between the chloride and amide moieties (rN−H···Cl = 3.266−
3.360 Å) were observed for all iridium pincer hydrido chloro
complexes containing the NH linker, 2e−h, in the solid state,
and this could play a dominant role in the structural distortion
observed. The influence of hydrogen bonding on structural
distortion is most profound for 2e. Unexpectedly different Ir−P
bond lengths, Ir−P1 = 2.2935(13) Å and Ir−P2 = 2.3186(13)
Å, were observed for this highly symmetrical complex. This can
(
tBu4POCOP)Ir(CO) (1937 cm−1) was observed in DCM, as
tBu is more electron donating than iPr.30 Both the literature and
current work demonstrate that the electronic properties of a
pincer-ligated metal center can be fine-tuned by systematic
altering the linker, the substituents at the phosphorus atom, or
both. Our attempts to prepare the corresponding carbonyl
complexes of 2g,h following the procedure in Scheme 5 were
unsuccessful (see the Supporting Information for details),
which prevents us from quantitatively measuring the π-back-
donation ability of (iPr2PNOCOPtBu2)Ir and (Cy2PNOCOPtBu2)-
Ir.
Molecular Structures of Iridium Pincer Complexes.
After a systematic spectroscopic study of the electronic
properties of newly synthesized iridium pincer complexes
2e−h, we turned our attention to the structural properties of
these iridium complexes.17 Good-quality crystals of iridium
complexes 2e−h were recrystallized from toluene and
characterized by X-ray crystallography (Figure 1).
E
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 3. Selected Bond Lengths and Bond Angles for Complexes 2e−h31
a
b
b
c
complex
linker
Ir−C1 (Å)
Ir−P1 (Å)
Ir−P2 (Å)
Ir−Cl1 (Å)
P1−Ir−P2 (deg)
C1−Ir−Cl1 (deg)
ref
2b
2e
2a
2f
O
2.010
2.297
2.293
2.404(1)
2.4245(13)
2.425
160.06
179.11(14)
177.22(13)
179.7(2)
14
NH
2.019(4)
2.014
2.2937(13)
2.305
2.3186(13)
2.305
163.44(5)
164.27
this work
10
CH2
NH/O
CH2/O
S/O
2.011(8)
2.016
2.278(2)
2.3194
2.310(2)
2.2685
2.401(3)
2.4012(10)
2.4149(8)
2.4199(12)
2.4127(9)
161.88(8)
163.55
178.8(3)
this work
12
2c
2d
2g
2h
177.71(9)
172.97(9)
173.91(14)
170.88(11)
2.019
2.2986
2.298
168.32
14
NH/O
2.011(5)
2.2797(12)
2.2904(12)
162.14(4)
162.20(3)
this work
this work
NH/O
2.013(3)
2.809(9)
2.2912(10)
a
b
c
Distance between Ir and C1. The bond lengths Ir−P1 and Ir−P2. Defined as P1∧IrP2.
be attributed to the fact that intermolecular hydrogen bonding
to the adjacent chloride ion only occurs at one amide moiety
(NH−P2) and not the other (NH−P1). The Cl···H−N
hydrogen bonding will reduce the electron density of the
nearest phosphorus atom P2, and consequently Ir−P2 is
elongated with respect to Ir−P1 in 2e. Closer inspection of the
solid-state structure of 2e shows that intermolecular hydrogen
bonding also increases the P2−Ir−C1 bond angle with respect
to the P1−Ir−C1 angle.
Despite the distorted-square-pyramidal structures owing to
intermolecular hydrogen bonding between chloride and NH
moieties, all four iridium pincer complexes containing one or
more NH linkers prepared in this work show a pronounced
planar structure: i.e., the arene backbone of the pincer ligand is
essentially coplanar with the two adjacent five-membered Ir−
P−X−C−C chelate rings. The coplanar structure of these
iridium pincer complexes could be essential for efficient π
conjugation between a lone pair of the NH/O linker and the
benzene backbone of the pincer ligand. The resulting stable and
rigid structure could confer increased thermal stability which is
highly desirable for dehydrogenation reactions.
The steric hindrance of transition-metal complexes is often
quantified by the P1−Ir−P2 bite angle, which is invariably
altered when different linkers are placed between the benzene
backbone and the phosphine moiety. As seen in Table 3 for the
series of tert-butyl-substituted phosphine-based iridium pincer
complexes, the P1−Ir−P2 bite angle of (tBu4PXCXP)IrHCl
increases with increasing covalent radii of the linker atoms:33
160.06° for 2b (X = O), 163.44° for 2e (X = NH), and 164.27°
for 2a (X = CH2). A similar trend was also observed for
unsymmetrical iridium pincer complexes, (tBu4PXCOP)IrHCl,
which exhibit increasing P1−Ir−P2 bite angles as the covalent
radii of the other linker increases: 161.88° for 2f (X = NH),
163.55° for 2c (X = CH2), and 168.32° for 2d (X = S).
Apparently the steric properties of the iridium pincer complex
can be effectively modulated by systematic altering of the
linkers and by their combinations. The structural parameters
determined above for iridium catalyst precursors in the solid
state may not be reflected in the real catalyst under the reaction
conditions. However, the structural dependence of iridium
pincer complexes on steric/electronic properties of the linkers
should follow a trend similar to that observed in the solid state
and provide a basis for further study of the structure−activity
relationship. Furthermore, the tendency of the NH moiety to
function as a strong hydrogen bond donor suggests that the
newly prepared iridium pincer complexes might potentially find
broad applications especially in catalytic activation and
transformation of polar covalent bonds, where effective
metal−ligand cooperation would be expected.
Catalytic Activity for Alkane Transfer Dehydrogen-
ation. Catalytic activities of 2e−h were evaluated for the
benchmark transfer dehydrogenation reaction of COA with
TBE as a hydrogen acceptor. The catalytic reactions were
carried out using a Fischer−Porter bottle with a total volume of
∼20 mL instead of the sealed small-volume glass reactor (∼4
mL) used in the literature;11 2e−h were activated in situ by
addition of NaOtBu under reaction conditions otherwise similar
to those employed in the literature.27 Each catalyst was tested
in triplicate, and the catalytic results of 2e−h are compared
with those of the reference catalysts 2a,b in Table 4.
Table 4. Catalytic Test Data for the Benchmark Transfer
a
Dehydrogenation Reaction Using COA and TBE
b
b
precatalyst
TON (6 h)
COE/(COE + 1,3-COD) (%)
2a
2b
2e
2f
1530
898
71.6
83.4
85.9
84.2
68.7
72.7
596
731
2g
2h
1627
1117
a
Reaction conditions: [Ir] = 1.1 mM, [COA] = [TBE] = 3.37 M, ∼10
b
equiv of NaOtBu, 200 °C, 6 h. Average of three or more runs.
All iridium complexes display considerable activities for
COA/TBE transfer dehydrogenation. In addition to cyclo-
octene (COE) as the expected product from transfer
dehydrogenation of COA, substantial amounts of 1,3-cyclo-
octadiene (1,3-COD) were also produced from consecutive
dehydrogenation of COA.11 As seen from Table 4, the COE
content in total products decreases with increasing catalyst
activity in the order 2e < 2f < 2b < 2h < 2a < 2g, since more
COE underwent further dehydrogenation to 1,3-COD over a
more active catalyst. Surprisingly, (tBuPCP)Ir (2a) was found to
be more active than (tBuPOCOP)Ir (2b) under the reaction
conditions employed in this study. This is completely opposite
to what has been reported in the literature: 2a is at least 8 times
less active than 2b under identical conditions,11 and the
catalytic activity of 2a is significantly inhibited when the ratio
TBE:Ir is >300.8 The surprisingly high activity of 2a for the
COA/TBE reaction in this study could be attributed to the
different experimental setup. The Fischer−Porter bottle reactor
used in this study has a large headspace; the ratio between
liquid phase and headspace volume is approximately 1:9 under
F
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
ambient conditions while the ratio was approximately 1:1 for
the sealed glass reactor mostly used in the literature.11,12 It is
therefore plausible to postulate that the concentration of TBE
in this study would be much lower than that in the literature
due to vapor−liquid equilibria; thus, 2a is much less inhibited
by olefins and displays activity comparable to that of 2b. The
catalytic effectiveness of iridium pincer catalysts can be
dramatically affected by reaction conditions in addition to
many other factors thoroughly reviewed in the literature, such
as electronic/steric properties of catalytically active species,17
the reaction mechanism,34 and the combination of alkane
substrate and hydrogen acceptor.1 The discrepancy in the
relative catalytic transfer dehydrogenations of 2a and 2b clearly
highlights challenges in a fair and accurate comparison of
catalyst performance.
of iridium pincer complexes and the electronic/steric effects of
pincer ligands for several reasons. It is predominantly because
the electronic effects cannot be completely separated from the
steric effects even with rationally designed and systematically
tuned ligand scaffolds. For instance, substitution of the
methylene linker for more electronegative O or NH linkers
increases the electron deficiency of the iridium metal center but
also results in considerable changes in the bond length to
neighboring atoms. Second, the catalytic activities were
determined at a single time point and may not reflect the
intrinsic activities of the catalyst, as potential catalyst
deactivation was largely neglected. Nevertheless, iridium pincer
complexes with an NH linker prepared in this study exhibit
activity comparable to or higher than those of their PCP and
POCOP analogues for the COA/TBE benchmark reaction.
The NH linker could play a greater role via metal−ligand
cooperation, a concept well demonstrated in other catalyst
systems with similar ligand scaffolds.35,36 The potential
applications of these novel iridium pincer complexes in other
transformations such as linear alkane dehydrogenation, alcohol
dehydrogenation, and CO2 hydrogenation are currently under
investigation.
Structural characterization and preliminary catalytic test
results in Tables 2−4 do not reveal a clear trend between
catalytic performance of iridium pincer complexes with
electronic/steric properties of varying linkers or alkyl
substituents at phosphorus. However, the results of 2b,e,f,
tert-butyl-substituted pincer iridium complexes with varying
linkers, reveal a weak correlation between catalytic activity and
the electronic/steric properties described by their C−O
stretching frequencies and P−Ir−P bite angles. Their COA/
TBE transfer dehydrogenation activities, 2e (TON = 596), 2f
(TON = 731), and 2b (TON = 898), increase with decreasing
P−Ir−P bite angle (2e, 163.44°; 2f, 161.88°; 2b, 160.06°) and
with increasing electron deficiencies of the iridium center (νCO
1919.5 cm−1 (3e), 1928.2 cm−1 (3f), 1937 cm−1 (3b)). Thus, a
smaller bite angle and more electron deficient metal center
seem to favor the COA/TBE reaction for this series of iridium
complexes. However, such a trend does not hold for
CONCLUSIONS
■
A simple modular synthetic route has been developed to
prepare unsymmetrical hybrid pincer ligands iPr2PNCOPtBu2
(1g) and Cy2PNCOPtBu2 (1h) in a one-pot process without
isolation and purification of the monosubstituted intermediate.
The series of iridium pincer complexes (tBu4PNCNP)IrHCl
(2e), (tBu4PNCOP)IrHCl (2f), (iPr2PNCOPtBu2)IrHCl (2g),
and (Cy2PNCOPtBu2)IrHCl (2h) have been synthesized and
fully characterized by 31P, H, and 13C NMR spectroscopy and
1
(
tBu4PCP)IrHCl (2a), which has the largest bite angle and
elemental analysis. Single-crystal X-ray crystallography studies
show that all iridium pincer complexes synthesized adopt a
slightly distorted square pyramidal geometry and exhibit
extensive hydrogen bonding between the NH functionality
with chloride ions of the adjacent iridium complex in the solid
state. The catalytic activities of iridium pincer complexes were
evaluated for the benchmark COA/TBE transfer dehydrogen-
ation and compared with those of tBuPCPIrHCl (2a) and
tBuPCPIrHCl (2b) reference catalysts. Iridium complexes with
NH linkers 2e−h exhibit moderate to high activities for COA/
TBE transfer dehydrogenation. (iPr2PNCOPtBu2)IrHCl (2g), an
iridium complex supported by unsymmetrical hybrid pincer
ligands with an NH linker, afforded the highest activity among
all iridium pincer complexes tested.
the most electron rich metal center but exhibits the highest
activity. We speculate that lone pairs of O and NH linkers could
considerably affect the electronic properties of the metal center
probably via effective overlapping of the p orbital of the linker
atoms with the ring current of benzene backbone, while the
overlapping of orbitals is insignificant for (tBu4PCP)IrHCl (2a).
For unsymmetrical (R2PNCOPtBu2)IrHCl complexes (R= tBu
i
(2f), Pr (2g), Cy (2h)), the steric properties of alkyl
substituents apparently play a dominant role in catalytic
activity. The COA/TBE reaction activity increases in the
order 2f (TON = 731) < 2h (TON = 1117) < 2g (TON =
1627), with the steric hindrance of alkyl substituents at the NH
arm decreasing from tBu to Cy to iPr. Interestingly, the P−Ir−P
bite angles of 2f−h determined by X-ray crystallography are
very similar: 2f, 161.88°; 2g, 162.14°; 2h, 162.20°. The catalyst
testing protocol used in this study determined the catalytic
activity at a single time point and does not take potential
catalyst deactivation into consideration, thus presenting an
obvious disadvantage for accurate comparison of catalyst
performance. Although no catalyst decomposition was noted
visually for any of the iridium catalysts during the course of the
reaction, a more thorough and reliable catalyst testing is
required to establish structure−activity relationships.
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were carried out
under a dry argon atmosphere, using an argon-filled glovebox or using
standard Schlenk and cannula techniques. DCM, THF, pentane, and
toluene were collected from a Braun solvent purification system and
deoxygenated by sparging with argon prior to use. C6D6 and CD2Cl2
were dried over P2O5 and vacuum-transferred prior to use. Resorcinol,
3-aminophenol, m-phenylenediamine, ClPtBu2, ClPiPr2, sodium
hydride, sodium tert-butoxide, n-butyllithium, P2O5, and [IrCl-
(COE)2]2 were purchased from Sigma-Aldrich and used without
further purification. Cyclooctane and 3,3-dimethyl-1-butene were
purchased from Sigma-Aldrich and degassed prior to use (the water
contents were determined to be 20 and 41 ppm, respectively, by Karl
Fischer titration). 1,3-Bis[(di-tert-butylphosphino)methyl]benzene
(1a) was purchased from STREM. NMR spectra were acquired on
Bruker AVIII 500 (500 MHz), Bruker AVIII HD-500 (500 MHz), and
Bruker AV 400 (400 MHz) spectrometers. Chemical shifts are
In summary, newly synthesized iridium pincer complexes
containing NH linkers 2e−h display moderate to high activities
for the benchmark COA/TBE transfer dehydrogenation
reaction; the highly unsymmetrical hybrid pincer complex
(
iPr2PNCOPtBu2)IrHCl (2g) was found to be the most active
among the iridium pincer complexes tested. It is difficult to
establish a clear correlation between the catalytic performance
G
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
referenced to residual protio impurities in the deuterated solvent (1H)
or the 13C shift of the solvent (13C). Phosphorus chemical shifts are
reported without reference. Mass spectra were obtained using a Bruker
electrospray ionization (ESI) micrOTOF-Q II mass spectrometer.
Elemental analysis was performed using a Thermo Scientific (Carlo
Erba) Flash 200 elemental analyzer. IR spectra were recorded on a
Nicolet Nexus FT-IR spectrometer.
(2.98 g) in quantitative yield. The purity was determined to be >90%
by 31P NMR, and the ligand was used without further purification.
1H NMR (CD2Cl2, 500.13 MHz, 300 K): δ 6.97 (apparent t, |2 ×
3JHH| = 8.1 Hz, 1H, Ar-H), 6.80 (m, 1H, Ar-H), 6.54 (m, 1H, Ar-H),
6.50 (m, 1H, Ar-H), 3.78 (d, 2JPH = 10.2 Hz, 1H, NH), 1.74 (m, 2H, P-
3
CH(CH3)2), 1.15 (d, JPH = 11.7 Hz, 18H, P-C(CH3)3), 1.08−1.02
(m, 12H, P-CH(CH3)2). 31P{1H} NMR (CD2Cl2, 202.46 MHz, 300
K): δ 152.5 (s), 48.4 (s). 13C{1H} NMR (CD2Cl2, 125.71 MHz, 300
Synthesis of Pincer Ligands. The ligand tBu4POCOP(1b) was
synthesized following literature procedures.11
2
2
K): δ 161.2 (d, JPC = 9.4 Hz, CAr), 150.6 (d, JPC = 16.4 Hz, CAr),
129.8 (s, CAr), 109.4 (d, 3JPC = 11.7 Hz, CAr), 108.6 (d, 3JPC = 10.6 Hz,
CAr), 106.2 (apparent t, |2 × 3JPC| = 11.5 Hz, CAr), 35.8 (d, 1JPC = 26.2
Hz, Cq), 28.0 (d, 2JPC = 17.0 Hz, CH3), 27.5 (d, 2JPc = 15.9 Hz, CH3),
tBu4PNCNP (1e).24 A solution of n-butyllithium (5.25 mL, 1.1 M in
hexane, 5.78 mmol) was slowly added via syringe to a solution of m-
phenylenediamine (272 mg, 2.5 mmol) in THF (20 mL) cooled to 0
°C. A yellow precipitate was formed; the mixture was stirred for 30
min at 0 °C and then for 3 h at room temperature. The solution was
then cooled to 0 °C, and ClPtBu2 (1.05 mL, 1.00 g, 5.5 mmol) was
added dropwise. The solution was warmed to room temperature and
then refluxed for 16 h. After evaporation of the solvent under vacuum,
the residue was extracted with pentane (3 × 20 mL) and the extract
cannula-filtered. The pentane was removed in vacuo to afford 1e (626
mg, 63% yield).
1
27.1 (d, 2JPC = 12.0 Hz, CH3), 19.0 (d, JPC = 19.9 Hz, CH), 17.2 (d,
1JPC = 7.8 Hz, CH).
Cy2PNCOPtBu2 (1h). From 3-aminophenol (546 g, 5 mmol), sodium
hydride (142 mg, 5.5 mmol), di-tert-butylchlorophosphine (1.1 mL,
5.5 mmol), triethylamine (1.4 mL, 10 mmol), and dicyclohexyl-
chlorophosphine (1.2 mL, 5.5 mmol) 1h was obtained as a white solid
(2.03 g, 90%). The purity was determined to be ∼95% by 31P NMR,
and the ligand was used without further purification.
3
1H NMR (CD2Cl2, 499.93 MHz): δ 6.91 (1H, t, JPH = 8.0, Ar-H),
1H NMR (C6D6, 499.93 MHz, 300 K): δ 7.23 (m, 1H, Ar-H), 7.09
(apparent t, |2 × 3JHH| = 8.2 Hz, Ar-H), 6.88 (m, 1H, Ar-H), 6.73 (m,
6.75 (1H, m, Ar-H), 6.38 (2H, dm, 3JPH = 8.0, Ar-H), 3.99 (2H, d, 2JPH
3
= 10.9, NH), 1.15 (36H, d, JPH = 11.8, PC(CH3)3). 31P{1H} NMR
2
1H, Ar-H), 3.45 (d, JPH = 10.7 Hz, 1H, NH), 1.73−1.51 (m, 12H,
(CD2Cl2, 125.76 MHz): δ 57.8 (s). 13C{1H} NMR (CD2Cl2,202.38
MHz): δ 151.0 (Cq, d, 2JPC = 17, C(1) and C(3)), 129.9 (CH, s, C(5)),
107.0 (CH, d, 3JPC = 11.6, C(4) and C(6)), 103.8 (CH, t, 3JPC = 12.5,
3
CyH), 1.40−1.32 (m, 2H, CyH), 1.17 (d, JPH = 11.6 Hz, 18H, P-
C(CH3)3), 1.15−1.01 (m, 8H, CyH). 31P{1H} NMR (C6D6, 202.38
MHz, 300 K): δ 150.3 (s), 41.0 (s). 13C{1H} NMR (C6D6, 125.76
MHz, 300 K): δ 161.4 (d, 2JPC = 9.7 Hz, CAr), 151.2 (d, 2JPC = 17.0 Hz,
1
2
C(2)), 34.6 (Cq, d, JPC = 20.0, C(CH3)3), 28.4 (CH3, d, JPC = 15.3,
C(CH3)3).
3
3
CAr), 130.1 (s, CAr), 109.8 (d, JPC = 12.5 Hz, CAr), 108.8 (d, JPC
=
tBu4PNCOP (1f). n-Butyllithium (5.5 mL, 1.1 M in hexanes, 6.05
mmol) was added dropwise to a solution of 3-aminophenol (273 mg,
2.5 mol) in THF (20 mL) at 0 °C followed by 1 h of stirring. The
solution was warmed to room temparture and stirred for a further 3 h,
3
11.7 Hz, CAr), 106.4 (apparent t, |2 × JPC| = 11.9 Hz, CAr), 36.7 (d,
1JPC = 13.1 Hz, CH), 35.7 (d, 1JPC = 26.3 Hz, Cq), 29.4 (d, 2JPC = 18.7
2
Hz, CH2), 27.6 (d, JPC = 15.7 Hz, CH3), 27.5−27.1 (m, CH2), 26.7
(s, CH2).
t
whereupon a copious off-white precipitate had formed. Bu2PCl (1.05
Synthesis of Iridium Complexes. tBuPCPIrHCl (2a) and
tBuPOCOPIrHCl (2b) were prepared by following the literature
procedure.11
mL, 5.5 mmol) was added dropwise to the resulting suspension at 0
°C. The reaction was warmed to room temperature, a condenser was
fitted, and the reaction mixture was brought to reflux for 15 h. The
volatile components were removed in vacuo, the resulting solid residue
was extracted with pentane (2 × 20 mL), and the extract was cannula-
filtered. The pentane was removed in vacuo to afford 1f as a brown
solid (452 mg, 45%), which displayed approximately 50% purity by 31P
NMR and was used without further purification.
General Procedure for Synthesis of Iridium Complexes 2e−h.
[IrCl(COE)2] and the ligands 1e−h (∼2.2 equiv) were dissolved in
toluene (15 mL) in a Schlenk flask, a condenser was fitted, and the
reaction mixture was refluxed overnight (or 4 h for 2h) and then
cooled to room temperature. The toluene was removed in vacuo, and
the solid residue was washed with pentane and dried in vacuo to
obtain 2e−h.
1H NMR (CD2Cl2, 500.13 MHz, 300 K): δ 6.98 (apparent t, |2 ×
3JHH| = 8.1 Hz, 1H, Ar-H), 6.83 (m, 1H, Ar-H), 6.58 (m, 1H, Ar-H),
tBu4PNCNPIrHCl (2e). From 1e (325 mg, 0.83 mmol) and
[IrCl(COE)2]2 (347 mg, 0.4 mmol) 2e was obtained as a brick red
solid (215 mg, 45% yield). The solid residue was washed with pentane
and then extracted into DCM. The DCM extract was taken to dryness
to afford the final product.
6.50 (m, 1H, Ar-H), 4.03 (d, 2JPH = 10.6 Hz, 1H, NH), 1.15 (d, 3JPH
=
3
11.7 Hz, 18H, P-C(CH3)3), 1.12 (d, JPH = 11.8 Hz, 18H, P-
C(CH3)3). 31P{1H} NMR (CD2Cl2, 202.46 MHz, 300 K): δ 152.4 (s),
58.7 (s). 13C{1H} NMR (CD2Cl2, 125.76 MHz, 300 K): δ 161.2 (d,
2
2JPC = 9.3 Hz, CAr), 151.3 (d, JPC = 17.6 Hz, CAr), 129.8 (s, CAr),
3
3
Anal. Calcd for C22H42ClIrN2P2: C, 42.33; H, 6.78; N, 4.49. Found:
C, 42.18; H, 6.91; N, 4.34. HRMS: calcd for C22H43ClIrN2P2 [M +
109.6 (d, JPC = 11.7 Hz, CAr), 108.5 (d, JPC = 10.8 Hz, CAr), 106.3
3
1
(apparent t, |2 × JPC| = 11.6 Hz, CAr), 35.8 (d, JPC = 25.7 Hz, Cq),
34.4 (apparent t, |2 × JPC| = 20.2 Hz, Cq), 28.5 (d, JPC = 15.0 Hz,
1
H+], 625.2206; found, m/z 625.2597. H NMR (C6D6, 500.13 MHz,
1
2
3
3
2
2
300 K): δ 6.92 (t, JHH = 7.7 Hz, 1H, Ar-H), 6.24 (d, JHH = 7.7 Hz,
CH3), 28.3 (d, JPC = 15.2 Hz, CH3), 27.5 (d, JPC = 15.4 Hz, CH3).
General Procedure for Synthesis of Ligands 1g,h. In an argon-
filled glovebox, 3-aminophenol and sodium hydride (1.1 equiv) were
weighed into a Schlenk flask inside a glovebox. THF (20 mL) was
added (caution! hydrogen evolution), and the resulting suspension was
refluxed for 1 h and then cooled to room temperature. Di-tert-
butylchlorophosphine (1.1 equiv) was added dropwise, and the
reaction mixture was refluxed for 1−1.5 h and then cooled to room
temperature. After completion of the reaction (indicated by 31P NMR
spectroscopy), triethylamine (2 equiv) and then either diisopropyl-
chlorophosphine or dicyclohexylchlorophosphine (1.1 equiv) were
added slowly at room temperature. The reaction mixture was refluxed
overnight. The volatiles were removed in vacuo, the residue was
extracted with pentane, and the extract was cannula-filtered. The
pentane was removed in vacuo to obtain the ligands 1g,h.
2H, Ar-H), 4.05 (br s, 2H, NH), 1.24 (apparent t, |2 × 3JPH| = 7.1 Hz,
3
18H, P-C(CH3)3), 1.19 (apparent t, |2 × JPH| = 7.2 Hz, 18H, P-
C(CH3)3), −41.99 (t, 2JPH = 13.0 Hz, 1H, IrH). 31P{1H} NMR (C6D6,
202.46 MHz, 300 K): δ 106.0 (s). 13C{1H} NMR (C6D6, 125.76 MHz,
300 K): δ 158.4 (apparent t, |2 × 2JPC| = 10.1 Hz, CAr), 125.1 (s, CAr),
4
1
101.5 (t, JPC = 5.7 Hz, CAr), 41.5 (apparent t, |2 × JPC| = 12.2 Hz,
Cq), 38.1 (apparent t, |2 × 1JPC| = 13.0 Hz, Cq), 28.4 (apparent t, |2 ×
2JPC| = 2.7 Hz, CH3), 28.3 (apparent t, |2 × JPC| = 2.7 Hz, CH3).
2
tBu4PNCOPIrHCl (2f). From 1f (428 mg, estimated purity 50%,
∼0.54 mmol) and [IrCl(COE)2]2 2f was obtained as a dark red solid
(145 mg, 55% yield).
Anal. Calcd for C22H41ClIrNOP2: C, 42.27; H, 6.61; N, 2.24.
Found: C, 42.19; H, 6.57; N, 2.33. HRMS: calcd for C22H42ClIrNOP2
1
[M + H+], 626.2046; found, m/z 626.2057. H NMR (C6D6, 500.13
MHz, 300 K): δ 6.86 (apparent t, |2 × 3JHH| = 7.8 Hz, 1H, Ar-H), 6.72
(d, 3JHH = 7.9 Hz, 1H, Ar-H), 6.25 (d, 3JHH = 7.6 Hz, 1H, Ar-H), 4.12
(d, 4JPH = 3.1 Hz, 1H, NH), 1.35 (d, 3JPH = 14.2 Hz, 9H, P-C(CH3)3),
1.29 (d, 3JPH = 14.5 Hz, 9H, P-C(CH3)3), 1.19 (d, 3JPH = 13.9 Hz, 9H,
iPr2PNCOPtBu2 (1g). From 3-aminophenol (763 mg, 7 mmol),
sodium hydride (194 mg, 8 mmol), di-tert-butylchlorophosphine (1.7
mL, 8 mmol), triethylamine (2.1 mL, 15 mmol), and diisopropyl-
chlorophosphine (1.3 mL, 8 mmol) 1g was obtained as a colorless oil
H
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3
P-C(CH3)3), 1.15 (d, JPH = 14.1 Hz, 9H, P-C(CH3)3), −41.34
Hz, 2H, Ar-H), 4.68 (br s, 2H, NH), 1.36 (apparent t, |3JPH| = 7.1 Hz,
(apparent t, |2 × JPH| = 13.1 Hz, 1H, IrH). 31P{1H} NMR (C6D6,
P-C(CH3)3). 31P{1H} NMR (CD2Cl2, 202.46 MHz, 300 K): δ 129.4
2
202.46 MHz, 300 K): δ 171.5 (d, 2JPP = 357 Hz), 108.5 (d, 2JPP = 357
Hz). 13C{1H} NMR (CD2Cl2, 125.76 MHz, 300 K): δ 168.4 (apparent
t, |2JPC + 3JPC| = 5.9 Hz, CAr), 158.5 (dd, 3JPC = 14.3 Hz, 2JPC = 5.4 Hz,
(s). 13C{1H} NMR (CD2Cl2, 125.76 MHz, 300 K): δ 199.7 (t, JPC
=
2
6.2 Hz, Ir-CO), 161.4 (apparent t, |2JPC + 3JPC| = 12.7 Hz, CAr), 128.2
(s, CAr), 99.9 (apparent t, |3JPC + 4JPC| = 6.5 Hz, CAr), 39.7 (apparent t,
3
3
|1JPC + JPC| = 13.2 Hz, Cq), 29.0 (apparent t, |2JPC + JPC| = 3.0 Hz,
3
3
CAr), 125.49 (s, CAr), 103.3 (d, JPC = 11.7 Hz, CAr), 102.5 (d, JPC
=
1
3
11.5 Hz, CAr), 43.3 (dd, JPC = 19.3 Hz, JPC = 5.0 Hz, Cq), 42.0 (dd,
1JPC = 19.6 Hz, 3JPC = 3.8 Hz, Cq), 39.6 (dd, 1JPC = 21.5 Hz, 3JPC = 5.5
Hz, Cq), 38.7 (dd, 1JPC = 21.8 Hz, 3JPC = 3.6 Hz, Cq), 28.4 (apparent t,
CH3).
tBu4PNCOPIr(CO) (3f). In a glovebox, tBuPNCOPtBuIrHCl (2f; 26
mg, 42 μmol) and sodium tert-butoxide (7 mg, 72 μmol) were
weighed into a Schlenk flask inside a glovebox. C6D6 (4 mL) was then
added under argon. The solution was stirred vigorously under
hydrogen for 2 h, during which time the solution turned from dark
red to orange. Despite the addition of further sodium tert-butoxide
(3.5 mg, 36 mmol), the reaction did not go to completion and the
conversion of 2f was 90%, as indicated by 31P{1H} NMR. The
hydrogen in the Schlenk flask was replaced with CO, and the solution
was stirred vigorously under CO, whereupon it rapidly turned yellow;
the reaction mixture was stirred under CO for a further 20 min. The
solution was transferred via cannula filtration into a separate Schlenk
flask, and the solvent was removed in vacuo to afford 3f as a yellow
solid showing over 90% purity by 31P NMR.
|2 × 2JPC| = 4.6 Hz, CH3), 27.9 (d, 2JPC = 4.8 Hz, CH3), 27.7 (d, 2JPC
=
4.9 Hz, CH3)
iPr2PNCOPtBu2IrHCl (2g). From 1g (1.7 mL, 0.3074 mM in toluene,
0.52 mmol) and [IrCl(COE)2]2 2g was obtained as a bright red solid
(193 mg, 65% yield).
Anal. Calcd for C20H37ClIrNOP2: C, 40.23; H, 6.25; N, 2.35.
Found: C, 40.38; H, 6.14; N, 2.38. HRMS: calcd for C20H38ClIrNOP2
[M + H+], 598.1733; found, m/z 598.1740. H NMR (C6D6, 499.93
1
MHz, 300 K): δ 6.87 (apparent t, |2 × 3JHH| = 7.9 Hz, 1H, Ar-H), 6.73
(d, 3JHH = 7.9 Hz, 1H, Ar-H), 6.22 (d, 3JHH = 7.7 Hz, 1H, Ar-H), 3.69
(d, 2JPH = 3.4 Hz, 1H, NH), 2.48 (m, 1H, P-CH(CH3)2), 1.95 (m, 1H,
P-CH(CH3)2), 1.33 (d, 3JPH = 14.0 Hz, 9H, P-C(CH3)3), 1.29 (d, 3JPH
= 14.5 Hz, 9H, P-C(CH3)3), 1.23−1.12 (m, 6H, P-CH(CH3)2), 0.87−
0.80 (m, 6H, P-CH(CH3)2), −39.79 (apparent t, |2 × 2JPH| = 13.2 Hz,
1H, IrH). 31P{1H} NMR (C6D6, 202.46 MHz, 300 K): δ 175.1 (d, 2JPP
HRMS: calcd for C23H41ClIrNO2P2 [M + H+], 618.2237; found, m/
1
z 618.2266. IR (DCM, cm−1): νCO 1928.2. H NMR (C6D6, 500.13
MHz, 300 K): δ 6.96 (apparent t, |2 × 3JHH|= 7.8 Hz, 1H, Ar-H), 6.82
= 362 Hz), 103.1 (d, JPP = 362 Hz). 13C{1H} NMR (C6D6, 125.76
(d, 3JHH = 8.0 Hz, 1H, Ar-H), 6.39 (d, 3JHH = 7.6 Hz, 1H, Ar-H), 4.38
3
MHz, 300 K): δ 168.0 (apparent t, |2JPC + JPC| = 5.4 Hz, CAr), 157.3
(d, JPH = 2.7 Hz, 1H, NH), 1.36 (d, JPH = 14.3 Hz, 18H, P-
C(CH3)3), 1.21 (d, 3JPH = 13.9 Hz, 18H, P-C(CH3)3). 31P{1H} NMR
(C6D6, 202.46 MHz, 300 K): δ 195.5 (d, 2JPP = 301 Hz), 128.9 (d, 2JPH
= 302 Hz). 13C{1H} NMR (C6D6, 125.76 MHz, 300 K): δ 199.8
3
2
3
(dd, 2JPC = 15.1 Hz, 3JPC = 4.6 Hz, CAr), 125.4 (s, CAr), 113.9 (s, CAr),
103.9 (d, 3JPC = 12.2 Hz, CAr), 103.2 (d, 3JPC = 11.8 Hz, CAr), 43.9 (dd,
1JPC = 18.5 Hz, 3JPC = 5.4 Hz, Cq), 39.3 (dd, 1JPC = 20.3 Hz, 3JPC = 5.4
(apparent t, |2 × 2JPC| = 5.5 Hz, Ir-CO), 129.0 (s, CAr), 102.6 (d, 3JPC
=
2
2
Hz, Cq), 28.8 (d, JPC = 3.3 Hz, CH3), 28.5 (d, JPC = 3.0 Hz, CH3),
13.2 Hz, CAr), 101.9 (d, 3JPC = 13.0 Hz, CAr), 41.0 (dd, 1JPC = 22.7 Hz,
3JPC = 3.1 Hz, Cq), 39.4 (d, 1JPC = 23.6 Hz, Cq), 28.9 (d, 2JPC = 6.0 Hz,
2
2
27.9 (d, JPC = 5.3 Hz, CH3), 27.7 (d, JPC = 5.5 Hz, CH3), 27.4 (d,
2JPC = 2.6 Hz, CH3), 17.8 (s, CH), 17.6 (m, CH), 16.8 (s, CH).
Cy2PNCOPtBu2IrHCl (2h). From 1h (242 mg, 0.5 mmol) and
[IrCl(COE)2] (197 mg, 0.22 mmol) 2h was obtained as a bright
red-pink solid (161 mg, 54% yield).
2
CH3), 28.6 (d, JPC = 6.2 Hz, CH3).
Crystallography. X-ray diffraction data for iridium hydrido chloro
complexes were collected at either 173 K (2e,f) or 93 K (2h) by using
a Rigaku FR-X Ultrahigh Brilliance Microfocus RA generator/confocal
optics and XtaLAB P200 system, with Mo Kα radiation (λ = 0.71075
Å). Diffraction data for compound 2g were collected at 173 K by using
a Rigaku MM-007HF High Brilliance RA generator/confocal optics
and XtaLAB P100 system, with Cu Kα radiation (λ = 1.54187 Å).
Intensity data were collected using either just ω steps or both ω and φ
steps, accumulating area detector images spanning at least a
hemisphere of reciprocal space. All data were corrected for
Lorentz−polarization effects. A multiscan absorption correction was
applied by using either CrystalClear37 or CrysAlisPro.38 Structures
were solved by Patterson (DIRDIF99 PATTY39) or dual-space
(SHELXT40) methods and refined by full-matrix least squares against
F2 (SHELXL-201641). Non-hydrogen atoms were refined anisotropi-
cally, and hydrogen atoms bound to carbon were refined using a riding
model. Hydrogen atoms bound to nitrogen or iridium were located
from the difference Fourier map and refined isotropically subject to a
distance restraint. All calculations were performed using the
CrystalStructure42 interface. Crystallographic data for the four
complexes have been deposited with the Cambridge Crystallographic
Data Centre as CCDC 1570250−1570253.
Anal. Calcd for C26H45ClIrNOP2: C, 46.11; H, 6.70; N, 2.07.
Found: C, 46.27; H, 6.86; N, 2.08. HRMS: calcd for C26H46ClIrNOP2
[M + H+], 678.2359; found, m/z 678.2362. H NMR (C6D6, 500.13
1
MHz, 300 K): δ 6.91 (apparent t, |2 × 3JHH| = 7.8 Hz, 1H, Ar-H), 6.75
(d, 3JHH = 7.8 Hz, 1H, Ar-H), 6.34 (d, 3JHH = 7.8 Hz, 1H, Ar-H), 3.93
(d, 2JPH = 2.7 Hz, 1H, NH), 2.50−2.40 (m, 2H, CyH), 2.,03−1.42 (m,
12 H, CyH), 1.35 (d, 3JPH = 14.2 Hz, 9H, P-C(CH3)3), 1.30 (d, 3JPH
=
14.4 Hz, 9H, P-C(CH3)3), 1.27−0.93 (m, 8H, CyH), −39.89
(apparent t, |2 × JPH| = 13.1 Hz, 1H, IrH). 31P{1H} NMR (C6D6,
2
2
2
202.46 MHz, 300 K): δ 175.2 (d, JPP = 361 Hz), 95.9 (d, JPP = 361
Hz). 13C{1H} NMR (C6D6, 125.76 MHz, 300 K): δ 168.0 (apparent t,
|2JPC + JPC| = 5.6 Hz, CAr), 157.6 (dd, JPC = 15.3 Hz, JPC = 5.7 Hz,
3
2
3
3
CAr), 125.4 (s, CAr), 114.2 (s, CAr), 103.9 (d, JPC = 12.2 Hz, CAr),
103.0 (d, 3JPC = 10.9 Hz, CAr), 43.8 (dd, 1JPC = 18.7 Hz, 3JPC = 5.3 Hz,
Cq), 39.3 (dd, 1JPC = 20.3 Hz, 3JPC = 5.6 Hz, Cq), 38.8 (dd, 1JPC = 29.3
Hz, 3JPC = 2.7 Hz, CH), 36.7 (dd, 1JPC = 30.9 Hz, 3JPC = 2.1 Hz, CH),
2
2
28.0 (s, CH2), 27.9 (d, JPC = 5.6 Hz, CH3), 27.7 (d, JPC = 5.1 Hz,
CH3), 27.4 (d, 2JPC = 3.4 Hz, CH2), 27.1 (s, CH2), 27.0 (d, 3JPC = 4.1
Hz, CH2) 26.9 (d, 4JPC = 4.8 Hz, CH2), 26.8 (d, 4JPC = 3.9 Hz, CH2),
26.2 (s, CH2), 26.1 (s, CH2).
tBu4PNCNPIr(CO) (3e). In a glovebox, 2e (24 mg, 40 μmol) and
sodium tert-butoxide (7 mg, 72 μmol) were weighed in to a Schlenk
flask inside a glovebox. C6D6 (2 mL) and COE (0.1 mL, 770 μmol)
were added, and the reaction mixture was heated to 90 °C for 1 h and
then cooled to room temperature. CO was then bubbled through the
solution, whereupon the color changed from dark orange to yellow;
the CO flow was maintained for 10 min to ensure completion of the
reaction. The solution was transferred into a separate Schlenk flask by
filtration, and the volatiles were removed in vacuo; the resulting orange
residue was washed with the minimum amount of pentane to afford a
yellow solid. 31P NMR analysis revealed quantitative conversion to the
carbonyl complex 3e.
General Procedure for the Transfer Dehydrogenation of
COA with TBE. Catalyst testing was conducted in a Fischer−Porter
bottle with a volume of ∼20 mL. The predried Fischer−Porter bottle
was charged with the required amount of sodium tert-butoxide and
then evacuated and back-filled with argon three times in a Schlenk line.
COA (1.2 mL) and TBE (1.15 mL) were added by syringe under
argon flow followed by a solution of the precatalyst (0.3 mL, 10 mM in
toluene, 3 μmol). The Fischer−Porter bottle was heated in a sand bath
of ∼1 cm above the liquid level. The apparatus was heated to 200 °C
for 6 h with a stirring rate of 400 rpm under an autogenous pressure of
∼2 bar. At the end of the reaction, the bottle was cooled with an ice−
water bath and the liquid product was analyzed by an Agilent 6850N
GC instrument equipped with a GS-GASPRO column (60 m × 0.32
mm × 0.5 μm). Oven temperature program: isothermal at 220 °C for
30 min, then 5 °C/min to 250 °C and hold at 250 °C for 30 min. The
HRMS: calcd for C23H42ClIrN2OP2 [M + H+], 617.2397; found, m/
z 617.2419. IR (DCM, cm−1): νCO 1919.5. 1H NMR (CD2Cl2, 499.93
MHz, 300 K): δ 6.62 (t, 3JHH = 7.7 Hz, 1H, Ar-H), 6.19 (d, 3JHH = 7.7
I
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
concentrations of hydrocarbon reactants and products were
determined using external standards.
(12) Nawara-Hultzsch, A. J.; Hackenberg, J. D.; Punji, B.; Supplee,
C.; Emge, T. J.; Bailey, B. C.; Schrock, R. R.; Brookhart, M.; Goldman,
A. S. ACS Catal. 2013, 3, 2505−2514.
(13) Yao, W.; Zhang, Y.; Jia, X.; Huang, Z. Angew. Chem., Int. Ed.
2014, 53, 1390−1394.
ASSOCIATED CONTENT
■
S
* Supporting Information
(14) Yao, W.; Jia, X.; Leng, X.; Goldman, A. S.; Brookhart, M.;
Huang, Z. Polyhedron 2016, 116, 12−19.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00713.
(15) Goldman, A. S.; Renkema, K. B.; Czerw, M.; Krogh-Jespersen,
K. ACS Symp. Ser. 2004, 885, 198−215.
(16) Kumar, A.; Bhatti, T. M.; Goldman, A. S. Chem. Rev. 2017, 117
(19), 12357−12384.
Synthesis and characterization of iridium hydrido chloro
carbonyl pincer complexes, ESI-MS of iridium pincer
complexes, crystallographic parameters for 2e,f,h, NMR
data for all compounds, preparation of 3e from 2e via
iridium hydride intermediate, and attempted preparation
of 3g,h (PDF)
(17) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A. S.
Chem. Rev. 2011, 111 (3), 1761−1779.
(18) Zhang, Y.; Yao, W.; Fang, H.; Hu, A.; Huang, Z. Sci. Bull. 2015,
60 (15), 1316−1331.
́
(19) Benito-Garagorri, D.; Bocokic, V.; Mereiter, K.; Kirchner, K.
Organometallics 2006, 25 (16), 3817−3823.
Accession Codes
(20) Murugesan, S.; Stoeger, B.; Weil, M.; Veiros, L. F.; Kirchner, K.
Organometallics 2015, 34, 1364−1372.
CCDC 1570250−1570253 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(21) Baroudi, A.; El-Hellani, A.; Bengali, A. A.; Goldman, A. S.;
Hasanayn, F. Inorg. Chem. 2014, 53, 12348−12359.
(22) Bordwell, F. G.; Branca, J. C.; Hughes, D. L.; Olmstead, W. N. J.
Org. Chem. 1980, 45 (16), 3305−3313.
(23) Gawron, O.; Duggan, M.; Grelecki, C. J. Anal. Chem. 1952, 24
(6), 969−970.
(24) Ozerov, O. V.; Guo, C.; Foxman, B. M. J. Organomet. Chem.
2006, 691 (22), 4802−4806.
AUTHOR INFORMATION
■
(25) Hrobar
́
ik, P.; Hrobar
́
ikova,
́
V.; Meier, F.; Repisky, M.;
́
Corresponding Author
*E-mail for J.L.: liujianke@hotmail.com.
Komorovsky, S.; Kaupp, M. J. Phys. Chem. A 2011, 115 (22), 5654−
́
5659.
ORCID
(26) Bistoni, G.; Rampino, S.; Scafuri, N.; Ciancaleoni, G.; Zuccaccia,
D.; Belpassi, L.; Tarantelli, F. Chem. Sci. 2016, 7 (2), 1174−1184.
(27) Goettker-Schnetmann, I.; White, P. S.; Brookhart, M.
Organometallics 2004, 23, 1766−1776.
(28) Huang, Z.; Brookhart, M.; Goldman, A. S.; Kundu, S.; Ray, A.;
Scott, S. L.; Vicente, B. C. Adv. Synth. Catal. 2009, 351, 188−206.
(29) Goldberg, J. M.; Cherry, S. D. T.; Guard, L. M.; Kaminsky, W.;
Goldberg, K. I.; Heinekey, D. M. Organometallics 2016, 35, 3546−
3556.
(30) Goldberg, J. M.; Wong, G. W.; Brastow, K. E.; Kaminsky, W.;
Goldberg, K. I.; Heinekey, D. M. Organometallics 2015, 34, 753−762.
(31) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41 (2), 466−470.
(32) Rimoldi, M.; Mezzetti, A. Inorg. Chem. 2014, 53, 11974−11984.
(33) Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.;
Echeverria, J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans.
2008, 21, 2832−2838.
(34) Renkema, K. B.; Kissin, Y. V.; Goldman, A. S. J. Am. Chem. Soc.
2003, 125, 7770−7771.
(35) Morello, G. R.; Hopmann, K. H. ACS Catal. 2017, 7, 5847−
5855.
(36) Tanaka, R.; Yamashita, M.; Nozaki, K. J. Am. Chem. Soc. 2009,
131, 14168−14169.
(37) CrystalClear-SM Expert v2.1; Rigaku Americas, The Woodlands,
TX, USA, and Rigaku Corporation, Tokyo, Japan, 2015.
(38) CrysAlisPro v1.171.38.43i and v. 1.171.39.8; Rigaku Oxford
Diffraction, Rigaku Corporation, Oxford, U.K., 2015.
(39) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garcia-Granda,
S.; Gould, R. O.; Israel, R.; Smits, J. M. M. DIRDIF-99;
Crystallography Laboratory, University of Nijmegen, Nijmegen, The
Netherlands, 1999.
David B. Cordes: 0000-0002-5366-9168
Jianke Liu: 0000-0002-4190-9195
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Sven Tobisch, Dr. Philip Dyer and Prof. David
Cole-Hamilton for helpful discussions. We also thank Sasol for
permission to publish this work.
REFERENCES
■
(1) Jensen, C. Chem. Commun. 1999, 24, 2443−2449.
(2) Kumar, A.; Goldman, A. S. In The Privileged Pincer-Metal
Platform: Coordination Chemistry & Applications; van Koten, G.,
Gossage, R. A., Eds.; Springer International Publishing: Cham,
Switzerland, 2016; pp 307−334.
(3) Gao, Y.; Guan, C.; Zhou, M.; Kumar, A.; Emge, T. J.; Wright, A.
M.; Goldberg, K. I.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem.
Soc. 2017, 139, 6338−6350.
(4) Roddick, D. M. In Organometallic Pincer Chemistry; van Koten,
G., Milstein, D., Eds.; Springer Berlin Heidelberg: Berlin, Heidelberg,
2013; Vol. 40, pp 49−88.
(5) Shih, W.-C.; Ozerov, O. V. Organometallics 2017, 36, 228−233.
(6) Polukeev, A. V.; Wendt, O. F. Organometallics 2017, 36, 639−
649.
(7) Stoutland, P. O.; Bergman, R. G.; Nolan, S. P.; Hoff, C. D.
Polyhedron 1988, 7 (16), 1429−1440.
(8) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M.
Chem. Commun. 1996, 17, 2083−2084.
(40) Sheldrick, G. Acta Crystallogr., Sect. A: Found. Adv. 2015, 71 (1),
3−8.
(9) Kundu, S.; Choliy, Y.; Zhuo, G.; Ahuja, R.; Emge, T. J.; Warmuth,
R.; Brookhart, M.; Krogh-Jespersen, K.; Goldman, A. S. Organo-
metallics 2009, 28, 5432−5444.
(41) Sheldrick, G. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71
(1), 3−8.
(10) Punji, B.; Emge, T. J.; Goldman, A. S. Organometallics 2010, 29,
2702−2709.
(42) CrystalStructure v4.2; Rigaku Americas, The Woodlands, TX,
USA, and Rigaku Corporation, Tokyo, Japan, 2015.
(11) Goettker-Schnetmann, I.; White, P.; Brookhart, M. J. Am. Chem.
Soc. 2004, 126, 1804−1811.
J
DOI: 10.1021/acs.organomet.7b00713
Organometallics XXXX, XXX, XXX−XXX